With the recent announcement of Ektachome returning to the market I decided to take and old expired roll of EliteChrome out of the freezer and run it through the Nikon F3. A friend asked if film is still better than digital and my response was, it’s different. A little better in some respects and not as good in others. The main talking points were highlight falloff, color rendering and pixels vs little globs of silver crystals. I tried to explain the bayer pattern of a digital sensor, but it’s easier to type than talk about over coffee so here it goes.
All digital sensors see black and white and the little pixels record brightness values. These black and white pixels are covered with red green and blue dots in a bayer pattern. This means that each pixel does not have complete R, G, and B information. Each color only covers one-third of the sensor. This translates into one-thirds of the resolution for each color, which means that the megapixel number of your camera is somewhat exaggerated.
Since each pixel only has one-third the color data needed to be resolved, digital cameras use something called the Bayer Interpolation in the tiny computer in the camera to interpolate, or guess the color values in between each pixel to come up with brightness value for any given color. So, the A7rII is 42 megapixels but in reality it only resolves about half of that and the rest is a result of the bayer interpolation algorithm, some guessing and smoothing.
So back to film. Film grain contains the full R, G and B resolution at every point and all of the color information and details are captured across the entire piece of film. So you get the maximum resolution for all the colors being captured.
Film resolution is also measured differently than megapixels. It’s measured in it’s ability to capture line pairs. A line pair comprises a dark line and an adjacent light line. A line is either a dark line or a light line. A resolution of 10 lines per millimeter means 5 dark lines alternating with 5 light lines, or 5 line pairs per millimeter
Let’s (safely) assume that Ektar resolves 100 line pairs per millimeter to make the math easy.
So, 100 pairs mean 200 lines, or 200 pixels per millimeter. With a little math we can figure a millimeter square of film is 200×200 or 40,000 pixels or 0.04 megapixels per square millimeter.
Not much right? Well when we’re talking about full frame which is 35mm x 24mm and with some math we get 864 square millimeters. More math and we multiply our 0.04 X 864 and we get a little bit less than the A7rII at 34.56 megapixels.
Here’s the kicker, our eyes don’t see in megapixels and each of those 34,560,000 bits of film grain has the full RGB information packed into it. It wasn’t interpolated by a tiny computer and it changes gradually from one tone to the next making it look more natural and more like our eyes function.
So is film better? Yeah, sort of, but it’s not all about megapixels or resolution. It wins because it’s organic and natural and sort of works like our eyes see things. I think they perfected film a long time ago and we’re still working to get digital sensors to catch up to where we left off with film. Film has a familiarity, we haven’t forgotten what it looked like when we picked up our glorious film prints at the drugstore.
We’re getting closer and digital sensors work really well but we’re a long way off from making film totally obsolete.